The beneficial effects of drugs in the treatment of HIV are particularly difficult to ascertain clinically in infected children and asymptomatic adults. Currently available laboratory tests to evaluate a drug's effectiveness, such as serum p24 quantitation, are very often negative in these two populations, or the test results may not change during therapy, e.g. isolation of virus from blood leukocytes. In the longitudinal study proposed here, HIV infected children and asymptomatic adults entering AZT therapy will be monitored with an immunoassay - ELISPOT. This assay, modified in our laboratory to enumerate peripheral blood lymphocytes which secrete HIV specific antibodies, has yielded positive results in about three-quarters of infected children or asymptomatic adults. In a preliminary cross-sectional survey, 12 of 13 adults receiving AZT were found to be ELISPOT negative for HIV. Since the number of antibody-secreting cells (ASC) is related to the extent of ongoing antigenic stimulation, the hypothesis to be tested is that, as AZT suppresses virus replication, the antigenic stimulus would be reduced, with a corresponding decrease in circulating ASC to HIV. We have also observed recently that the total number of circulating B cells secreting immunoglobulin of many isotypes is greatly elevated in asymptomatically infected adults, as compared to non-infected individuals. Since this finding is likely to be related to a direct or indirect effect of HIV in non-specific B cell activation, a reduction in virus replication would be expected to affect the total number of circulating immunoglobulin-secreting-cells (IgSC). In the proposed longitudinal study, patients will serve as their own control. The number of circulating HIV-ASC or IgSC determined at two different times prior to administration of the drug will be compared to the results obtained at various intervals after initiation of therapy. Comparisons of measurements made after discontinuation (and possibly subsequent reinstitution) of treatment should also be helpful in assessing the effects of the drug. Although not the primary objective of this study, comparisons to other laboratory tests and to clinical manifestations may provide additional information on the usefulness of the ELISPOT methodology in determining the effectiveness of drugs of potential value for the treatment of HIV infection in children and adults.